Processing of vibrometer signals for determination of cardiovascular parameters

Laser Doppler vibrometers (LDV) measure tiny vibration amplitudes and have already shown to be suitable for many applications in technical and biomedical fields as, for example, vibration detection on the skin. The monitoring of the heart rhythm is actually based on the electrocardiogram (ECG), which requires the application of electrodes. A non-contact measurement has many benefits and would simplify patient monitoring (e.g., burned skin, chest-trauma, premature infants, and athletes during exercise). As already known, optical Vibrocardiography (VCG) with LDV allows the detection of heart rate and heart rate variability because the heart beat is an mechanical source. This paper demonstrates the possibility of a reliable detection and classification of atrioventricular (AV) blocks. To solve this task we have identified an area on the thorax which shows vibration responses of the contraction of the atrium and the ventricle. We also discuss different signal processing concepts for an automated signal pattern recognition of characteristic signal segment.